HummingBird is an engineering visualizer primarily aimed at making visualization of streaming data simple and effective. 

In addition to visualization, data correction and KPI generation are important capabilities of HummingBird

The product architecture is flexible, configurable and extensible, making it possible to consume data from a wide variety of data sources such as WITSML, WITS, Modbus, OPC/UA and many other protocols. 

Key Performance Indicators generation​

The key features of HummingBird are listed below :

  • Scroll graph – The live scroll graph view allows visualization of multiple data channels, streaming at different frequencies, on a single page. User is offered the flexibility of selecting data channels and customizing visual representation by assigning colors as per algorithmic rules.
  • LogViews – display data for the chosen channels of the entire well. The view also features a bit vs hole-depth difference graph along the time axis. The user can zoom-in and inspect the curves. Also included are ‘Remarks’ entered by the drilling operator.
  • KPIs – Basic Operational Events and complex Operational Events are processed by HummingBird into KPIs. These KPIs are presented graphically with various selection criteria. In-built KPIs include operational event analysis, connection analysis and torque and drag analysis. A framework to build customized KPIs is in place for users to develop their ow n KPIs.
  • Data Wrangling – HummingBird allow s depth and event correction. Markers can be added for additional information during correction. Data gaps are identified and user is given a choice about its usage. Approval mechanism is in place for the corrections and audit trail is maintained.
  • Adaptors and Operators – Various adaptors and operators can be configured to integrate external data sources and apply filters, such as decimation and simple moving average, to the data while being displayed graphically. A user with administrator privileges can configure adaptors and operators.
  • Dashboard Designer – A versatile graphical tool helps users design customized dashboards. The user can choose, place, size and align different dashboard components such as graphs, gauges, dials, etc. The user can also choose f rom a list of mnemonics to display and/or control.
  • Units of Measurement – HummingBird allow s users to select the displayed units which may differ from the source UoM.
  • Control Widgets – HummingBird is not only a data visualizer but is also a control interface that allows users, with the right privilege, to send set points and other advisory messages back to the rig’s operating system via Yellow Hammer. Support for device ‘TARE’ operation is also provided.
  • Administration – By integrating with LDAP servers such as Microsoft Active Directory, administrators can exercise control on data exposure at a well (or asset) level or at a predefined dashboard level.

The next steps

𝗣𝗿𝗲𝗱𝗶𝗰𝘁𝗶𝘃𝗲 𝗠𝗮𝗶𝗻𝘁𝗲𝗻𝗮𝗻𝗰𝗲(𝗣𝗱𝗠)

Imagine traditional maintenance like rigid, one-size-fits-all production lines. OEM-mandated schedules are like fixed assembly lines, reliable but potentially inflexible. They ensure essential checks are done but might not address individual equipment needs.

Reactive maintenance? That’s like waiting for a machine to malfunction before intervention – cheap upfront, but risky business in the long run. Preventive maintenance is like regular quality checks: good for catching issues early, but sometimes like endless inspections when specific diagnostics are needed. It can miss problems and waste resources.

Now, 𝗲𝗻𝘁𝗲𝗿 𝘁𝗵𝗲 𝗴𝗮𝗺𝗲-𝗰𝗵𝗮𝗻𝗴𝗲𝗿: 𝗣𝗱𝗠! Imagine your equipment whispering its health secrets through sensors – tiny microphones picking up vibrations, pressure and temperature changes, and other clues. . Our powerful AI/ML models act like expert diagnosticians, analyzing this data in real-time to predict potential issues before they turn into full-blown breakdowns. It’s like having a customized maintenance plan designed just for your equipment, optimizing interventions based on its unique data and usage fingerprint.

While traditional methods have their place, PdM offers several advantages:

  • Reduced downtime
  • Optimized maintenance
  • Extended equipment life
  • Data-driven decisions
  • Better planning

Of course, implementing PdM requires investment in sensors, data infrastructure, and expertise. But it’s a worthwhile investment for a personalized diagnostic system that offers all the benefits above.

However, it’s crucial to remember that PdM isn’t a silver bullet. It works best in conjunction with existing practices. Think of it as collaborating with an experienced technician: the OEM schedule provides a solid foundation, and PdM personalizes the approach for maximum efficiency. And the best approach depends on your specific context and resources.

AI/ML-powered PdM opens exciting possibilities for tailored maintenance, optimized resources, accurate failure predictions, and has the potential to revolutionize industries. Let’s leverage it to the fullest.